1. Field of the Invention
The present invention relates to an error correction device and an optical disk device, and more particularly to error correction of data which is expressed using phase modulation.
2. Description of Related Art
HD (High-Definition) DVDs have been proposed in recent years as next generation DVDs. It is intended that the HD-DVDs adopt the disk structure of the current DVDs so as to ensure compatibility therewith, while simultaneously achieving higher density data recording than that of the previous generation DVDs.
As one main feature, HD-DVDs adopt the land/groove recording method in which information is recorded in both lands and grooves. Land tracks and groove tracks are formed in wobbles, in which address information is embedded. More specifically, address information is expressed using phase modulation in such a manner that four waves having a phase of 0 degrees represent “0” and four waves having a phase of 180 degrees represent “1”. Here, a plurality of, in this case, four, waves with the same phase are provided in order to increase the accuracy of address information detection by means of redundancy. For the address information to be embedded in the form of wobbles, binary data are converted into a gray code in which the distance between codes (inter-code distance) for adjacent sets of binary data, that is, the number of inverted bits, is equal to 1. Accordingly, in gray code, address “0” is represented as “00000000”; address “1” is represented as “00000001”; address “2” is repented as “00000011”; address “3” is represented as “00000010”; address “4” is represented as “00000110”, and so on.
On the other hand, it is also possible that the address information is embedded by forming grooves in wobbles and data is recorded or reproduced with respect to only grooves. In this case, as it is not necessary to provide the address information to lands, the need for converting the address information into gray code before being embedded is eliminated. It is also possible to add check bits for CRC (Cyclic Redundancy Check) or CRC bits to the address information of the grooves. When data is recorded only in grooves, the address information is also expressed using phase modulation in such a manner that four waves having a phase of 0 degrees represent bit data “0” and four waves having a phase of 180 degrees represent bit data “1”, in the same manner as in the land/groove recording. Here, four waves having identical phases are used to express one-bit data both in an optical disk in which data is recorded in grooves and lands and in an optical disk in which data is recorded in grooves only, in order to allow both types of optical disks to be driven by a single optical disk device. As such, it is possible to enable a single disk device to drive both “HD DVD-R” (write-once or Recordable) which is an optical disk in which data is recorded in grooves only and “HD DVD-RW” (Rewritable) which is an optical disk in which data is recorded in grooves and lands. Such HD-DVDs are generally described in the Oct. 13, 2003 issue of “Nikkei Electronics” (Nikkei B P, Oct. 13, 2003, pp. 126-134).
As described above, the address information is expressed by means of phase modulation in such a manner that four waves having a phase of 0 degrees represent a bit value “0” and four waves having a phase of 180 degrees represent a bit value of “1”, and the bit value of “0” or “1” can be determined basically according to the principle of majority rule. More specifically, when three waves having a phase of 0 degrees and only one wave having a phase of 180 degrees are detected for one bit, the bit value can be determined to be “0”.
However, even when three waves having a phase of 0 degrees are detected, there is no guaranteeing that the bit data is truly “0”. It is therefore desirable to further increase the detection accuracy. Further, when two waves having a phase of 0 degrees and two waves having a phase of 180 degrees are detected, i.e. when a half of the four waves represent 0 and the other half of the four waves represent 1, it is not possible to determine which of the data 0 or 1 is represented by such wobbles. Of course, a reading error bit can be determined by performing CRC using CRC bits. With the CRC technique, however, even when 9-bit check bits are added, for example, it is only possible to detect an error in three or less bits. Accordingly, with the CRC technique, it is not possible to correct an error and determine a correct address.